Streaming audio over a packet-switched network, such as the Internet, typically involves a client device receiving a portion of the audio, buffering that portion for playout, and playing out the buffered portion while receiving another portion of the audio. Streaming is typically triggered by a user, for instance, when the user selects particular audio content and instructs an application to begin streaming the selected content. Doing so, however, requires the user's attention, and does not account for environments in which the user desires to listen to the audio, but cannot easily perform the selection thereof.